What impact do clinical histories have on radiologists' diagnostic performance?
How does the availability of a thorough clinical history impact radiologists' diagnostic performance?
A recent study published in Academic Radiology suggests that, in most cases, such knowledge improves performance. But there are instances in which it can result in positive disease diagnoses without clinical correlation.
“Having access to a patient's clinical information prior to diagnostic image interpretation has been a debated topic since the 1960s,” corresponding author Kehn E. Yapp, BDSc, of the Medical Image Optimization and Perception Group (MIOPeG) and Faculty of Medicine and Health at the University of Sydney in Australia, and co-authors explained. “This may be due to the cognitive biases that available history has on subsequent image interpretation, such as anchoring bias – the tendency to focus on salient features in the initial presentation without adjusting this initial impression in light of later information.”
Studies that have investigated the correlation between clinical histories and reader accuracy and/or diagnostic performance have uncovered mixed findings, with some suggesting that radiologists benefit from the added information and others finding that it created bias, which consequently led to an increase in false positive rates. The authors note the most recent systemic review on the effects of clinical history was completed in 2004. Since then, new methods of measuring diagnostic accuracy have been implemented, which is what prompted researchers to have another look at the data.
“The uncertainty and variation in results across studies that have explored the effect of clinical history on diagnostic performance emphasize the need for a review of the literature on this topic,” the authors wrote.
The experts searched through Medline, Embase, Scopus, Web of Science and the Cochrane Central Register of Controlled Trials (CENTRAL) databases to find studies that compared the diagnostic reader performance of radiologists interpreting exams both with and without available clinical histories. This resulted in a total of 22 studies that met inclusion criteria.
Of those studies, 15 demonstrated an improvement in diagnostic performance when clinical histories were available. No significant performance changes were observed in six studies, and one reported a decrease in performance. Two of the studies found that clinical histories increased location sensitivity but had no overall impact on performance.
“The clinical history, in a similar way to disease prevalence, establishes and updates the pretest probability — an assessment of how likely a patient has a specific disease,” the researchers said. “These data are combined with the information from the imaging to form the posterior (post-test) probability and is a likely explanation for greater diagnostic performance when imaging is read with clinical history.”
The experts noted the variations of study design for the publications included in their analysis and suggested that further studies are needed to better understand the significance of their findings.
“These should reflect the task in clinical practice, allow for measuring multiple abnormalities per image, reward correct and penalize incorrect abnormality locations, account for memory bias and allow adequate time between reading sessions, provide a balanced reading design and include the entire spectrum of disease and nondisease severity in the study population,” the authors explained.
More on radiology practice management:
Viewing medical images encourages patients to adhere to healthier lifestyles, meta-analysis shows
World imaging leaders share 7 strategies to optimize CT dosage
Breast density notifications leave many women in the dark, survey results suggest
Transitioning from peer reviews to peer learning: Benefits, challenges and implementation
In addition to her background in journalism, Hannah also has patient-facing experience in clinical settings, having spent more than 12 years working as a registered rad tech. She joined Innovate Healthcare in 2021 and has since put her unique expertise to use in her editorial role with Health Imaging.